Automated SON Rule Creation via Data Mining Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current Self-Organizing Networks (SON) algorithms rely on pre-analysis and human intervention, making it time-consuming to detect and implement new rules, especially with the rise in network elements, and there is a lack of automated mechanisms to convert data mining insights into actionable network control instructions.
Innovation Solution
A system and method using data mining to generate or adapt SON rules by analyzing 'Big Data' from radio access networks, comprising a data analysis component, rule creation engine, and element management system with rule translation and execution engines, automatically creating or adapting rules and converting them into event-condition-action parameters for network configuration changes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If pre-analysis and human intervention are used to create SON rules, then rule accuracy and reliability are improved, but rule creation time and operational complexity increase significantly
Solution Approach 1:
The system enables automated rule creation through data mining algorithms that autonomously analyze network data, identify patterns, and generate SON rules without human intervention. The rule creation engine automatically processes measurement data, detects correlations, and formulates configuration rules, allowing the system to serve itself rather than relying on manual engineering processes.
Solution Approach 2:
The patent replaces the manual mechanical process of rule creation with an automated information processing system. Data mining algorithms substitute human analysts, automatically extracting insights from network data and converting them into actionable SON rules through computational analysis rather than human reasoning and documentation.
2Productivity
If more network elements are added to handle increasing network size, then network capacity is improved, but the complexity of manual rule management and analysis increases
Solution Approach 1:
The automated rule creation system continuously monitors network elements and self-adjusts configuration rules based on real-time data analysis. The system autonomously adapts to network growth by automatically processing data from new network elements and generating appropriate rules without requiring proportional increases in manual management resources.
Solution Approach 2:
The data mining-based rule creation engine serves multiple network elements simultaneously through a universal analysis framework. A single automated system handles rule generation for diverse network elements (base stations, core network components, transmission elements) using unified data processing algorithms, eliminating the need for separate manual analysis processes for each element type.
3Measurement precision
If data mining is used to analyze Big Data from the network, then identification of event correlations and network behavior patterns is improved, but the automation capability to convert insights into actionable rules is insufficient
Solution Approach 1:
The patent merges the data mining analysis function with the rule creation function into a unified automated system. The rule creation engine directly integrates with the data mining process, automatically converting detected patterns and correlations into executable SON rules without manual intervention. This combines analytical precision with automated rule generation in a single continuous process.
Solution Approach 2:
The rule creation engine acts as an intermediary between data mining insights and network control actions. It automatically translates raw analytical results into standardized rule formats that can be directly applied to network elements, bridging the gap between data analysis and actionable automation through systematic rule formulation and validation processes.
4Adaptability or versatility
If manual processes are used to define and implement SON rules, then adaptability to new network scenarios is limited, but the system remains easier to control and validate
Solution Approach 1:
The system implements dynamic rule creation that automatically adapts to changing network conditions and new scenarios. The data mining engine continuously analyzes evolving network data patterns, and the rule creation engine dynamically generates updated rules without requiring manual reconfiguration. This enables the system to naturally adapt to new network scenarios, traffic patterns, and failure modes through continuous automated learning and rule generation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Data mined in a radio access and transport network is used to create or adapt SON rules and SON parameters. More particularly, the analytics of "Big Data" mined in the network are used to generate new or modified SON rules and/or parameters, in realtime, in an automated way (i.e., substantially without human interaction in creating/updating the rules and/or parameters). A system and method for creating or adapting a rule in a SON network, is provided including a first network management layer (which can be, but does not have to be, the network management layer of a "Big Data" system), an element management system layer, and at least one network element. In this embodiment, the system is configured to: obtain data mined from the SON network and/or other sources; perform analytics on the mined data; and automatically create a new rule or adapt an existing rule, based on the results of the analytics performed.